AB-620 study guide
Work through the domains in weight order: Integrate, then Plan, then Manage. Drill each table until every scenario cue points to one answer. When two options both seem to work, the right one is usually the governed, lowest-risk choice.
Know cold vs. look up later
| Know Cold | Searchable If Needed |
|---|---|
| Domain weights: INTEGRATE 40–45, PLAN 30–35, MANAGE 20–25 | Exact Settings page paths |
| Generative vs classic orchestration: what changes | Full list of 11 trigger names |
| Teams/M365 → Authenticate with Microsoft only | Manual-auth field names |
| Which integration fits which need (knowledge / tool / MCP / computer use / agent) | Wizard step labels |
| MCP = tools + resources, generative only | MCP transport details |
| Child vs connected agent criteria; Fabric data agent limits | Preview/GA status of each agent type |
| Knowledge limits: 25 web/SharePoint (gen), 4 (classic); Dataverse 2×15 classic | Semantic index file-size limits |
| Allow ungrounded responses OFF → uncited answer → Fallback | Citation rendering per channel |
| Evaluation: 8 graders and when to use each | Content-safety level numbers beyond 0 = strictest |
| ALM: custom solution → env vars → managed → pipelines into Managed Environments | Pipeline host setup screens |
Integrate and extend agents (40–45%)
Which integration?
| Need | Answer | Trap |
|---|---|---|
| Ground in third-party content indexed by Microsoft Search | Copilot connector knowledge (user's Entra identity) | It's indexed, so it isn't "live" |
| Answer from live LOB data, no index | Power Platform connector as real-time knowledge | Copilot connector = index |
| Write/update a record | Connector tool (or REST/MCP tool) | Knowledge never writes |
| Server publishing tools that change over time | MCP (tools + resources; needs generative) | Prompts aren't supported |
| API with an OpenAPI spec | REST API tool (v2 JSON; v3 auto-converted) | Always lands in a solution |
| No API, only a UI | Computer use (5 credits/step; 15 premium) | Default = maker credentials |
| Another agent on another framework | A2A (endpoint URL; card at /.well-known/agent.json) | Not the card URL |
| Foundry agent | Project endpoint + Agent ID; Activity endpoint must be enabled (else 400) | Old-portal agents → 404 |
| Fabric data agent | Add it as an agent | No redirect, no / reference, no M365 Copilot, no evaluation |
| Same team, group tools | Child agent | — |
| Other team, own ALM/model, reuse | Connected agent (~30–40 choices is the split point) | A main agent with connected agents can't itself be connected |
Knowledge know-cold
- Web 25 / SharePoint 25 / Dataverse & connectors unlimited in generative mode. Classic: 4 / 4 per node / 2×15 / 2.
- More than 25 sources → GPT filtering by description; uploaded files don't count.
- Bing Custom Search, custom data, Azure OpenAI "on your data" → Classic data in a generative answers node.
- Azure AI Search: integrated vectorization; one index per connection; Entra (service principal / integrated) over keys; citations need a URL field (
metadata_storage_path); VNet via private endpoint + Power Platform VNet support. - Official source isn't compatible with generative orchestration.
Azure monitoring
- Settings › Advanced › Application Insights connection string. Options: logging, conversation details (user + text), sensitive properties, node execution events.
- KQL
customEvents; exclude tests withdesignMode == "False". Block the feature with a data policy on the Application Insights in Copilot Studio connector.
Plan and configure agent solutions (30–35%)
Orchestration
| Behavior | Generative (default) | Classic |
|---|---|---|
| Topic selection | Name + description | Trigger phrases |
| Tools | Agent picks by description | Only from topics |
| Knowledge | Proactive | Fallback / node |
| Multi-intent | Chains several | One topic |
| Missing inputs | Agent asks | Question nodes |
| Conversational boosting edits | Ignored | Used |
| Multiple Topics Matched | Not called | Used |
| Custom entities as inputs | Not supported | n/a |
Controls: End all topics cancels the plan; Reset Conversation clears globals, not history (use Clear variable values › conversation history).
Identity and channels
- Teams + M365 → Authenticate with Microsoft only; others block those channels.
- Need
User.AccessTokenor a custom site with SSO → Authenticate manually. - No auth = anyone with the link. Sharing-based access control needs Entra (not generic OAuth2).
- Auth changes apply after publish. D365 Customer Service handoff → no Authenticate with Microsoft.
Agent flows
- Tool-callable only with When an agent calls the flow; return values with Respond to the agent.
- Billing: Copilot Studio capacity. From a topic = 1 classic answer + actions; from the orchestrator = 1 autonomous action + actions; test chat is free.
- Cloud flow → agent flow: in a solution, one-way.
- Multistage approvals: AI stages + manual stages + conditions; no ALM, no sharing; "Analysis failed" = contradictory instructions.
Topic nodes
- HTTP node: headers carry auth; default 30 s timeout; Continue on error → status code + body into variables.
- Generative answers node: Search only selected sources (no fallback); clear Send a message to post-process into an Adaptive Card; node moderation beats agent moderation.
- Variables: receive vs return; Date/time, duration, choice and custom-entity values can't be passed; global → topic conversion is one-way; secrets = env var of type secret → Key Vault.
- Entities: regex is case-sensitive (
(?i)); smart matching for fuzzy closed lists; One of multiple entities takes only the first match.
Test and manage agents (20–25%)
| Need | Grader |
|---|---|
| No single right answer | General quality |
| Same meaning, any wording | Compare meaning (default pass 50) |
| Exact code/number | Exact match |
| Near-identical wording | Text similarity |
| Right tool/topic used | Tool use |
| Required terms | Keyword match (Any/All) |
| Harm check | Content safety (0 = strictest) |
| Business rule, or refusal counts as a pass | Custom (instructions + labels) |
- Missing expected answer → Invalid. User profile simulates a signed-in user. Automate via REST API or connector. Fabric data agents aren't supported.
- ALM: default → custom solution (set preferred solution); env vars per target; managed into test/prod; pipeline targets must be Managed Environments; delegated deployments for approval + service principal; no cross-tenant; no skipping stages; import = Upgrade.
Habits to unlearn from older courses
| Older material shows | Current Copilot Studio |
|---|---|
| Typing "trigger phrases" into the topic trigger | A one–two sentence description of what the topic does and doesn't do |
| "Copilot Studio uses Azure CLU" | Generative orchestrator picks topics, tools, knowledge and agents by description |
| Escalate shows a message | Transfer conversation node → engagement hub, va_* context variables |
| Three trigger types | 11 types plus priority order (activity → message/event/update/invoke → agent chooses/phrases) |
| GPT-4.1 / Sonnet 4.5 / Opus 4.5 lineup | Default GPT-5.5 Chat; learn the Deep / General / Auto categories |
| Graph connectors | Copilot connectors |
Traps that look right
| Tempting answer | Why it's wrong |
|---|---|
| "Add more trigger phrases" to fix topic confusion | Generative mode → rewrite descriptions to be distinct |
| "Enable Multiple Topics Matched" | Not called under generative orchestration |
| "Edit Conversational boosting" to change knowledge search | Ignored in generative mode |
| "Reset Conversation" to clear planner memory | Clears globals only |
| "Add a line to instructions requiring an approved model" | Instructions don't select models; prompt model settings do |
| Copilot connector for "live, no indexing" | Copilot connectors are indexing |
| Add the Fabric data agent via Redirect | Unsupported |
| Deploy unmanaged to production via pipelines | Pipelines only deploy managed to targets |
| General quality for refusal tests | Penalizes abstention; use Custom |
| Web channel security for internal-only access | Secures the embed, not user identity; use Entra auth |